Estimating Chinook Salmon Passage in the Kenai River Using Dual Frequency Identification Sonar

Fixed location, side-looking split-beam sonar has been used to assess Chinook salmon Oncorhynchus tshawytscha passage into the Kenai River since 1995.  Estimating Chinook salmon passage is complicated by the presence of smaller more abundant species of salmon.  Classifying acoustic targets by species remains a difficult task and introduces uncertainty in estimates of Chinook salmon abundance largely because of the difficulty in accurately estimating the size of migrating fish.  Currently we use acoustic parameters based on echo shape to estimate fish size.  This study investigates whether dual-frequency identification sonar (DIDSON) can provide improved estimates of fish size that could lead to better methods for apportioning acoustic targets by species.   In this study, fish lengths are manually measured from DIDSON still images of migrating fish.  Length frequency and age data are also taken from fish captured by gillnets drifted onsite.  A species¨Mage mixture model can be fitted to the data to provide estimates of Chinook salmon passage.  Additionally, passage estimates for larger Chinook salmon can be produced using a simple threshold applied to DIDSON length measurements.